Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We’re working on 1M Covid-19 testing capacity per day

billiontoone.com

Post reply on HN
Print view View original

191–200 of 247 posts

Re: We’re working on 1M Covid-19 testing capacity per day

#191

Post by douglaswlance »

rplnt wrote: post #166
limaoscarjuliet wrote: post #162

Germany decided to mix 10 specimens and run a test, only then - if positive - test individual samples. They increased their testing capacity 10x (roughly) overnight. https://edition.cnn.com/world/live-news/coronavirus-pandemic...

There was a study, IIRC from Israel, which said this is still reliable with 64 samples mixed. The next question is, do you need to test the individual samples then? Not sure how much available material for testing you have, but you might be able to just divide the mix to eliminate bigger groups first. Or even mix parts with new samples. There must be an ideal procedure (throughput-wise) if you know how the range of p…

You can do a binary search to efficiently locate the positive sample.

Re: We’re working on 1M Covid-19 testing capacity per day

#192

Post by jacquesm »

panpanna wrote: post #188
tirant wrote: post #177

Earlier quoted context omitted.

I don't see Spanish politicians making more or less mistakes that other governments in Europe. Spain was one of the first countries to adopt strict measures to limit mobility of its citizens. And according to Google Mobility Analysis [1] Spaniards have mostly complied with it, staying at home at much higher rates than other neighbouring countries. So no, I don't see a reason to say that its politicians are a joke and…

Spains problem could at least partially be attributed to really bad luck (e.g. a Champions League game against an Italian team)

At least they cancelled the Fallas just in time or the disaster would have been much bigger.

Re: We’re working on 1M Covid-19 testing capacity per day

#193

Post by thehappypm »

shushpanchik wrote: post #133

I'm wondering, if the bottleneck in testing is number of tests per day available, could we use Bloom filter methodology for it? Like, for example, take samples of 1024 people, assign them 10-bit IDs randomly, mix samples of everyone with bit 1 in position 0 in one pool, with bit 1 in position 1 in 2nd pool and so on. Then do 10 tests, and whoever has negative result in any of set bits of his ID, does not have virus.…

I have heard many false negatives are an issue. Diluting samples down might make that even worse.

Re: We’re working on 1M Covid-19 testing capacity per day

#194

Post by yters »

phaemon wrote: post #182
yters wrote: post #82

Earlier quoted context omitted.

Yes, if you google "coronavirus iceland" you'll see they've tested almost their entire population. They found 1586 confirmed cases, of whom only 6 died, which gives a fatality rate of 0.4%. Turning this around, if we multiply confirmed deaths by 264, that gives us an estimate of how many cases there are. So, for example, with UK's death count of 6159 this means about 2.4% of the population is infected. Furthermore, o…

From your own link: "The failings contributed to the worst flu season for seven years, with 15,000 deaths from the virus, around twice the average figure, and the worst NHS performance on record." Not 50,000.

Sorry, I misread. This doc says (look at last couple pages) the 2014/2015 and 2017/2018 seasons had over 25k deaths from flu.

https://assets.publishing.service.gov.uk/government/uploads/...

So my calculation shows cov2 is about twice as bad as a strong flu, with no preventative measures. If we hit the 20k or less with preventative measures that the Imperial model predicts, this will be on par with regular flu season.

At any rate, much less than the 200k+ deaths originally predicted by the Imperial model and that pushed the UK to lockdown.

Re: We’re working on 1M Covid-19 testing capacity per day

#195

Post by mhandley »

Is the 1M/day figure for individual tests? When the expected number of positives is low, can't you run group tests? Mix the first samples from 16 patients together and test them as a batch. If the result is negative, all 16 are negative. Otherwise re-run the second samples. Depending on how many samples you take from each patient, you can binary search, or just run 16 individual tests on the second samples. By doing this, you can get something upwards of 10x increase in people tested per day.

Re: We’re working on 1M Covid-19 testing capacity per day

#196

Post by yters »

JamisonM wrote: post #98
yters wrote: post #82

Earlier quoted context omitted.

Yes, if you google "coronavirus iceland" you'll see they've tested almost their entire population. They found 1586 confirmed cases, of whom only 6 died, which gives a fatality rate of 0.4%. Turning this around, if we multiply confirmed deaths by 264, that gives us an estimate of how many cases there are. So, for example, with UK's death count of 6159 this means about 2.4% of the population is infected. Furthermore, o…

Iceland has only tested 28,992 of its residents as April 6th, so that throws the rest of this conjecture out the window. (Although regardless the conjecture was probably not a great idea.) https://covid.is/data

The point is Iceland has the best statistic on death rate due to comprehensive testing.

Re: We’re working on 1M Covid-19 testing capacity per day

#197

Post by ced »

douglaswlance wrote: post #191
rplnt wrote: post #166

Earlier quoted context omitted.

There was a study, IIRC from Israel, which said this is still reliable with 64 samples mixed. The next question is, do you need to test the individual samples then? Not sure how much available material for testing you have, but you might be able to just divide the mix to eliminate bigger groups first. Or even mix parts with new samples. There must be an ideal procedure (throughput-wise) if you know how the range of p…

You can do a binary search to efficiently locate the positive sample.

But what if there are two (or more?) positive samples? It's an interesting problem.

Re: We’re working on 1M Covid-19 testing capacity per day

#198

Post by Robotbeat »

platyp wrote: post #148
Robotbeat wrote: post #131

Earlier quoted context omitted.

South Korea shows that test and trace works just as good or better than lockdowns. Which is why our failure to get high testing rates early on is such an utter travesty.

South Korea does not "tell" us any of that. The government did not force a lockdown, but massive voluntary social distancing occurred anyway.

That’s not a lockdown, though. Voluntary behavior change is much better (than having to force a lockdown) if you can manage it.

Re: We’re working on 1M Covid-19 testing capacity per day

#199

Post by phreeza »

limaoscarjuliet wrote: post #162

Germany decided to mix 10 specimens and run a test, only then - if positive - test individual samples. They increased their testing capacity 10x (roughly) overnight. https://edition.cnn.com/world/live-news/coronavirus-pandemic...

https://en.wikipedia.org/wiki/Group_testing

Re: We’re working on 1M Covid-19 testing capacity per day

#200

Post by bequanna »

ced wrote: post #197
douglaswlance wrote: post #191

Earlier quoted context omitted.

You can do a binary search to efficiently locate the positive sample.

But what if there are two (or more?) positive samples? It's an interesting problem.

If the initial test size is 64 and contains a positive, split into two groups of 32 and test each. Continue splitting and testing when a positive is encountered.
Post reply on HN